Methods, systems, and apparatus, including computer programs encoded on computer storage media, for processing an input sequence using a recurrent neural network to generate an output for the input sequence. One of the methods includes receiving the input sequence; generating a doubled sequence comprising a first instance of the input sequence followed by a second instance of the input sequence; and processing the doubled sequence using the recurrent neural network to generate the output for the input sequence.
Legal claims defining the scope of protection, as filed with the USPTO.
1. A method for processing an input sequence using a recurrent neural network to generate an output for the input sequence, the method comprising: receiving the input sequence; generating a doubled sequence comprising a first instance of the input sequence followed by a second instance of the input sequence, wherein generating the doubled sequence comprises concatenating the first instance of the input sequence and the second instance of the input sequence; and processing the doubled sequence using the recurrent neural network to generate the output for the input sequence.
2. The method of claim 1 , wherein generating the doubled sequence comprises inserting a separator character between the first instance and the second instance so that the first instance is separated by the separator character from the second instance in the doubled sequence.
3. The method of claim 1 , wherein generating the doubled sequence comprises: reversing at least one of the first instance or the second instance in the doubled sequence.
4. The method of claim 1 , wherein the recurrent neural network is a long short term (LSTM) neural network.
5. The method of claim 1 , wherein the input sequence is a training sequence that is associated with a known output, wherein the recurrent neural network has a plurality of parameters, and wherein the method further comprises: training the recurrent neural network on the training sequence by adjusting values of the parameters of the recurrent neural network based on an error between the output generated by the recurrent neural network from the doubled sequence and the known output for the training sequence.
6. The method of claim 1 , wherein the input sequence is a sequence of characters from code of a computer program, wherein the doubled sequence comprises a first instance of the characters from the code followed by a second instance, and wherein the output generated by the recurrent neural network from the doubled sequence is a predicted output of the computer program.
7. A system comprising one or more computers and one or more storage devices storing instructions that when executed by the one or more computers cause the one or more computers to perform operations for processing an input sequence using a recurrent neural network to generate an output for the input sequence, the operations comprising: receiving the input sequence; generating a doubled sequence comprising a first instance of the input sequence followed by a second instance of the input sequence, wherein generating the doubled sequence comprises concatenating the first instance of the input sequence and the second instance of the input sequence; and processing the doubled sequence using the recurrent neural network to generate the output for the input sequence.
8. The system of claim 7 , wherein generating the doubled sequence comprises inserting a separator character between the first instance and the second instance so that the first instance is separated by the separator character from the second instance in the doubled sequence.
9. The system of claim 7 , wherein generating the doubled sequence comprises: reversing at least one of the first instance or the second instance in the doubled sequence.
10. The system of claim 7 , wherein the recurrent neural network is a long short term (LSTM) neural network.
11. The system of claim 7 , wherein the input sequence is a training sequence that is associated with a known output, wherein the recurrent neural network has a plurality of parameters, and wherein the method further comprises: training the recurrent neural network on the training sequence by adjusting values of the parameters of the recurrent neural network based on an error between the output generated by the recurrent neural network from the doubled sequence and the known output for the training sequence.
12. The system of claim 7 , wherein the input sequence is a sequence of characters from code of a computer program, wherein the doubled sequence comprises a first instance of the characters from the code followed by a second instance, and wherein the output generated by the recurrent neural network from the doubled sequence is a predicted output of the computer program.
13. A computer program product encoded on one or more non-transitory storage media, the computer program product comprising instructions that when executed by one or more computers cause the one or more computers to perform operations for processing an input sequence using a recurrent neural network to generate an output for the input sequence, the operations comprising: receiving the input sequence; generating a doubled sequence comprising a first instance of the input sequence followed by a second instance of the input sequence, wherein generating the doubled sequence comprises concatenating the first instance of the input sequence and the second instance of the input sequence; and processing the doubled sequence using the recurrent neural network to generate the output for the input sequence.
14. The computer program product of claim 13 , wherein generating the doubled sequence comprises inserting a separator character between the first instance and the second instance so that the first instance is separated by the separator character from the second instance in the doubled sequence.
15. The computer program product of claim 13 , wherein generating the doubled sequence comprises: reversing at least one of the first instance or the second instance in the doubled sequence.
16. The computer program product of claim 13 , wherein the recurrent neural network is a long short term (LSTM) neural network.
17. The computer program product of claim 13 , wherein the input sequence is a training sequence that is associated with a known output, wherein the recurrent neural network has a plurality of parameters, and wherein the method further comprises: training the recurrent neural network on the training sequence by adjusting values of the parameters of the recurrent neural network based on an error between the output generated by the recurrent neural network from the doubled sequence and the known output for the training sequence.
18. The computer program product of claim 13 , wherein the input sequence is a sequence of characters from code of a computer program, wherein the doubled sequence comprises a first instance of the characters from the code followed by a second instance, and wherein the output generated by the recurrent neural network from the doubled sequence is a predicted output of the computer program.
Cooperative Patent Classification codes for this invention. Click any code to explore related patents in that topic.
October 7, 2015
May 19, 2020
Browse 5M+ US patents with plain-English claim translations and AI-generated analysis.